Skip to content

perf tools: Add support to accumulate hist periods (v4)

Namhyung Kim lkml 1 version In review

Get series with b4
b4 am [email protected]

Versions newest first

  1. v1 Latest21 patches 31 repliesIn review

Patches in v1 No applies recorded yet

#Subject and filesSizeRepliesApplied
1/21perf tools: Introduce struct add_entry_itertools/perf/builtin-report.c+278 -907 replies
2/21perf hists: Convert hist entry functions to use struct he_stattools/perf/util/hist.c+11 -111 reply
3/21perf hists: Add support for accumulated stat of hist entrytools/perf/util/hist.c tools/perf/util/sort.h tools/perf/util/symbol.h+20 -05 replies
4/21perf hists: Check if accumulated when adding a hist entrytools/perf/builtin-annotate.c tools/perf/builtin-diff.c tools/perf/builtin-report.c +4 more+26 -170 replies
5/21perf hists: Accumulate hist entry stat based on the callchaintools/perf/builtin-report.c tools/perf/ui/stdio/hist.c+103 -23 replies
6/21perf tools: Update cpumode for each cumulative entrytools/perf/builtin-report.c+31 -32 replies
7/21perf report: Cache cumulative callchainstools/perf/builtin-report.c+48 -01 reply
8/21perf hists: Sort hist entries by accumulated periodtools/perf/builtin-report.c tools/perf/util/hist.c+18 -00 replies
9/21perf ui/hist: Add support to accumulated hist stattools/perf/ui/hist.c tools/perf/util/hist.h+46 -03 replies
10/21perf ui/browser: Add support to accumulated hist stattools/perf/ui/browsers/hists.c+17 -01 reply
11/21perf ui/gtk: Add support to accumulated hist stattools/perf/ui/gtk/hists.c+16 -01 reply
12/21perf tools: Apply percent-limit to cumulative percentagetools/perf/ui/browsers/hists.c tools/perf/ui/gtk/hists.c tools/perf/ui/stdio/hist.c+44 -122 replies
13/21perf tools: Add more hpp helper functionstools/perf/ui/hist.c tools/perf/util/hist.h+20 -00 replies
14/21perf report: Add --children optiontools/perf/Documentation/perf-report.txt tools/perf/builtin-report.c+15 -00 replies
15/21perf report: Add report.children config optiontools/perf/builtin-report.c+4 -00 replies
16/21perf tools: Factor out sample__resolve_callchain()tools/perf/builtin-report.c tools/perf/builtin-top.c tools/perf/util/callchain.c +1 more+38 -351 reply
17/21perf tools: Factor out fill_callchain_info()tools/perf/builtin-report.c tools/perf/util/callchain.c tools/perf/util/callchain.h+46 -370 replies
18/21perf top: Support callchain accumulationtools/perf/builtin-top.c+103 -32 replies
19/21perf top: Add --children optiontools/perf/Documentation/perf-top.txt tools/perf/builtin-top.c+13 -00 replies
20/21perf top: Add top.children config optiontools/perf/builtin-top.c+12 -00 replies
21/21perf tools: Enable --children option by defaulttools/perf/util/symbol.c+6 -52 replies

Reviews trailers per patch

2 of 21 patches have a Reviewed-by or Acked-by.

PatchIngo MolnarJiri Olsa
1/21
2/21A-b
3/21
4/21
5/21
6/21
7/21
8/21
9/21
10/21
11/21
12/21
13/21
14/21
15/21
16/21A-b
17/21
18/21
19/21
20/21
21/21Requested-by

Trailers given in replies are filled; those carried in the patches from earlier versions are outlined with dashes.

Read the v1 thread

Cover letter [PATCHSET 00/21] perf tools: Add support to accumulate hist periods (v4)

Hello,

This is my third attempt to implement cumulative hist period report. This work begins from Arun's SORT_INCLUSIVE patch [1] but I completely rewrote it from scratch.

Please see the patch 01/21. I refactored functions that add hist entries with struct add_entry_iter. While I converted all functions carefully, it'd be better anyone can test and confirm that I didn't mess up something - especially for branch stack and mem stuff.

This patchset basically adds period in a sample to every node in the callchain. A hist_entry now has an additional fields to keep the cumulative period if --children option is given on perf report.

I changed the option as a separate --children and added a new "Children" column (and renamed the default "Overhead" column into "Self"). The output will be sorted by children (cumulative) overhead for now. The reason I changed to the --children is that I still think it's much different from other --callchain options and I plan to add support for showing (remaining) callchains to cumulative entries too as Arun requested. The --callchain option will take care of it even with --children option.

I know that the UI should be changed also to be more flexible as Ingo requested, but I'd like to do this first and then move to work on the next. I also added a new config option to enable it by default.

* changes in v4:
- change to --children option (Ingo)
- rebased on new annotation change (Arnaldo)
- support perf top also
- enable --children option by default (Ingo)

* changes in v3:
- change to --cumulate option
- fix a couple of bugs (Jiri, Rodrigo)
- rename some help functions (Arnaldo)
- cache previous hist entries rathen than just symbol and dso
- add some preparatory cleanups
- add report.cumulate config option

Let me show you an example:

  $ cat abc.c
  #define barrier() asm volatile("" ::: "memory")
  void a(void)
  {
  	int i;
  	for (i = 0; i < 1000000; i++)
  		barrier();
  }
  void b(void)
  {
  	a();
  }
  void c(void)
  {
  	b();
  }
  int main(void)
  {
  	c();
  	return 0;
  }

With this simple program I ran perf record and report:

  $ perf record -g -e cycles:u ./abc
  $ perf report --stdio
      88.29%      abc  abc                [.] a                  
                  |
                  --- a
                      b
                      c
                      main
                      __libc_start_main
       9.43%      abc  ld-2.17.so         [.] _dl_relocate_object
                  |
                  --- _dl_relocate_object
                      dl_main
                      _dl_sysdep_start

       2.27%      abc  [kernel.kallsyms]  [k] page_fault         
                  |
                  --- page_fault
                     |          
                     |--95.94%-- _dl_sysdep_start
                     |          _dl_start_user
                     |          
                      --4.06%-- _start

       0.00%      abc  ld-2.17.so         [.] _start             
                  |
                  --- _start

When the -g cumulative option is given, it'll be shown like this:

  $ perf report --children --stdio
  #     Self  Children  Command      Shared Object                   Symbol
  # ........  ........  .......  .................  .......................
  #
       0.00%    88.29%      abc  libc-2.17.so       [.] __libc_start_main  
       0.00%    88.29%      abc  abc                [.] main               
       0.00%    88.29%      abc  abc                [.] c                  
       0.00%    88.29%      abc  abc                [.] b                  
      88.29%    88.29%      abc  abc                [.] a                  
       0.00%    11.61%      abc  ld-2.17.so         [.] _dl_sysdep_start   
       0.00%     9.43%      abc  ld-2.17.so         [.] dl_main            
       9.43%     9.43%      abc  ld-2.17.so         [.] _dl_relocate_object
       2.27%     2.27%      abc  [kernel.kallsyms]  [k] page_fault         
       0.00%     2.18%      abc  ld-2.17.so         [.] _dl_start_user     
       0.00%     0.10%      abc  ld-2.17.so         [.] _start             

As you can see __libc_start_main -> main -> c -> b -> a callchain show up in the output.

I know it have some rough edges or even bugs, but I really want to release it and get reviews. It does not handle event groups and annotations yet.

You can also get this series on 'perf/cumulate-v4' branch in my tree at:

git://git.kernel.org/pub/scm/linux/kernel/git/namhyung/linux-perf.git

Any comments are welcome, thanks.
Namhyung

Cc Arun Sharma <[email protected]>
Cc Frederic Weisbecker <[email protected]>

[1] https://lkml.org/lkml/2012/3/31/6

Namhyung Kim (21):
perf tools: Introduce struct add_entry_iter
perf hists: Convert hist entry functions to use struct he_stat perf hists: Add support for accumulated stat of hist entry perf hists: Check if accumulated when adding a hist entry perf hists: Accumulate hist entry stat based on the callchain perf tools: Update cpumode for each cumulative entry perf report: Cache cumulative callchains
perf hists: Sort hist entries by accumulated period perf ui/hist: Add support to accumulated hist stat perf ui/browser: Add support to accumulated hist stat perf ui/gtk: Add support to accumulated hist stat perf tools: Apply percent-limit to cumulative percentage perf tools: Add more hpp helper functions
perf report: Add --children option
perf report: Add report.children config option
perf tools: Factor out sample__resolve_callchain() perf tools: Factor out fill_callchain_info()
perf top: Support callchain accumulation
perf top: Add --children option
perf top: Add top.children config option
perf tools: Enable --children option by default

 tools/perf/Documentation/perf-report.txt |   5 +
 tools/perf/Documentation/perf-top.txt    |   6 +
 tools/perf/builtin-annotate.c            |   3 +-
 tools/perf/builtin-diff.c                |   2 +-
 tools/perf/builtin-report.c              | 534 +++++++++++++++++++++++++------
 tools/perf/builtin-top.c                 | 137 +++++++-
 tools/perf/tests/hists_link.c            |   4 +-
 tools/perf/ui/browsers/hists.c           |  51 ++-
 tools/perf/ui/gtk/hists.c                |  27 +-
 tools/perf/ui/hist.c                     |  62 ++++
 tools/perf/ui/stdio/hist.c               |  13 +-
 tools/perf/util/callchain.c              |  65 ++++
 tools/perf/util/callchain.h              |   8 +
 tools/perf/util/hist.c                   |  73 +++--
 tools/perf/util/hist.h                   |   7 +-
 tools/perf/util/sort.h                   |   1 +
 tools/perf/util/symbol.c                 |  11 +-
 tools/perf/util/symbol.h                 |   1 +
 18 files changed, 855 insertions(+), 155 deletions(-)
Signature · 1 line
1.7.11.7